Multiclass Kidney Abnormalities Detecting Novel System Through Computed Tomography

Accurate and early detection of kidney abnormalities is essential for effective clinical diagnosis and prevention of severe renal complications. Computed Tomography (CT) imaging is widely used for kidney examination due to its high resolution; however, traditional diagnosis relies heavily on manual interpretation by radiologists, which is time-consuming, subjective, and prone to human error. Although deep learning techniques have been applied to kidney disease detection, many existing systems are limited to binary classification, suffer from poor performance on low-contrast or noisy CT images, and lack robustness for multi-class abnormality detection. To address these challenges, this paper presents a Multi-Class Kidney Abnormalities Detecting Novel System through Computed Tomography using deep learning techniques. The proposed framework integrates CT image preprocessing, kidney segmentation, hybrid convolutional neural network (CNN)–based feature extraction, feature fusion, and multi-class classification. Advanced preprocessing techniques such as normalization, noise removal, contrast enhancement, and data augmentation are applied to improve image quality. A U-Net–based segmentation module isolates kidney regions for precise analysis, while hybrid CNN architectures extract both global and local features for effective representation. The system classifies CT images into multiple categories including normal kidney, kidney stone, renal cyst, and tumour. Experimental evaluation demonstrates high classification accuracy, reliable class-wise performance, and strong robustness across varying CT image conditions. Compared to traditional handcrafted feature-based and binary classification approaches, the proposed system achieves improved diagnostic accuracy with reduced manual intervention. The developed framework provides an efficient, automated, and reliable solution for multi-class kidney abnormality detection, supporting radiologists in early diagnosis and clinical decision-making.